Journal of Experimental Psychology: General1989, VoL 118, No.1, 100-104

Copyright 1989 by the American Psychological Association. Inc,0096-3445/89/$00,75

Misinformation and Memory:The Creation of New Memories

Elizabeth F. Loftus and Hunter G. HoffmanUniversity of Washington

Misleading information presented after an event can lead people to erroneous reports of thatmisinformation. Different process histories can be responsible for the same erroneous report indifferent people. We argue that the relative proportion of times that the different process historiesare responsible for erroneous reporting will depend on the conditions of acquisition, retention,and retrieval of information. Given the conditions typical of most misinformation experiments,it appears that misinformation acceptance plays a major role, memory impairment plays somerole, and pure guessing plays little or no role. Moreover, we argue that misinformation acceptancehas not received the appreciation that it deserves as a phenomenon worthy of our sustainedinvestigation. It may not tell us anything about impairment of memories, but it does tell ussomething about the creation of new memories.

Once upon a time, a man (whom we'll call Mike) stumbledupon an armed robbery in a hardware store. The robberrummaged around the cluttered store brandishing a silverweapon; finally, he stole all the money. Then, almost as anafterthought, he grabbed a hand calculator and a hammer,placing these in his satchel as he left the store. The police weresummoned immediately, but before they arrived, Mike talkedto another customer about the robbery. We'll call her Maria.Maria told Mike that she saw the robber grab a calculator anda screwdriver, stuffing them in his satchel as he left the store.The police arrived, and when they questioned Mike, he re­counted the robbery at some length: He described in detailthe silver weapon, the money, and the calculator. When thepolice asked him about a tool that they heard had been taken,"Did you see if it was a hammer or a screwdriver?", he said,"Screwdriver."

How did it happen that an ordinary upstanding guy likeMike came to remember seeing a screwdriver? (a) He mightnever have seen the hammer in the first place, and he men­tioned the screwdriver because he remembered hearing aboutit. (b) He could have remembered both the hammer and thescrewdriver, but he mentioned the screwdriver when askedbecause he trusted Maria's memory more than he trusted hisown. (c) He could have failed to see the hammer and failedto hear Maria mention the screwdriver, and he simply guessedabout the tool when asked by the police. Last, (d) he couldhave initially had a memory for a hammer, but when Mariamentioned the screwdriver, his memory was altered, sup­pressed, or impaired in some way. In fact, if there had beenfour customers in Mike's shoes that day, they might have allreported seeing a screwdriver, each for a different reason. Put

The underlying research fundamental to the arguments made herewas supported by a grant from the National Institute of MentalHealth.

We thank G. Loftus for his helpful comments.Correspondence concerning this article should be addressed to

Elizabeth F. Loftus, Department of Psychology, University of Wash­ington, Seattle, Washington 98195.

100

another way, entirely different process histories can lead tothe same final response.

Mike's erroneous report is analogous to the thousands oferroneous reports after the receipt of misinformation thathave been obtained in laboratory studies of the "misinfor­mation effect" conducted in the United States, Canada, GreatBritain, Germany, Australia, and the Netherlands (Ceci, Ross,& Toglia, 1987a, 1987b; Ceci, Toglia, & Ross, 1988; Chandler,1989; Geiselman, 1988; Gibling & Davies, 1988; Gudjonsson,1986; Hammersley & Read, 1986, in press; Kohnken &Brockmann, 1987; Kroll & Ogawa, 1988; Kroll & Timourian,1986; Lehnert, Robertson, & Black, 1984; Morton, Hammers­ley, & Bekerian, 1985; Pirolli & Mitterer, 1984; Register &Kihlstrom, 1988; Sheehan, 1988; Sheehan & Tilden, 1986;Smith & Ellsworth, 1987; Wagenaar & Boer, 1987; Zaragoza& Koshmider, 1989; Zaragoza, McCloskey, & Jamis, 1987).This enthusiasm for investigating the misinformation effecthas been fueled by an abiding interest on the part of research­ers in uncovering the mechanism that produces it.

When the first collection of misinformation experimentsappeared in the mid-1970s, the lesson that was being learnedfrom these experiments was that misleading postevent infor­mation can impair memory ofan original event (Loftus, 1975,1977, 1979). According to the "impairment" view, Mike'srecollection ofa screwdriver came about because of the fourthmechanism just cited: that his memory for the hammer hadbeen altered by the misleading postevent information. Thenotion of memory alteration bothered people. It challengedthe prevailing textbook view that memories, once stored, arepermanently stored; that traces once formed always survive;and that forgetting is due to a labile retrieval system (Atkinson& Shiffrin, 1968; Chechile, 1987). In fact, at least one theoryexplicitly claimed that postevent information does not impairunderlying memory traces; rather, it impairs only accessibilityto those original memories (Morton et aI., 1985).

Among the more articulate of those who were bothered byall notions of impairment were McCloskey and Zaragoza(1985). They claimed that memory for an original event isnot impaired by misleading postevent information. According

COMMENTS 101

to the "no-impairment view," Mike's recollection of a screw­driver came about because of one of the first three mecha­nisms cited earlier: that he never saw the hammer but remem­bered hearing about the screwdriver, that he rememberedboth the hammer and the screwdriver but trusted the screw­driver information more, or that he remembered neither toolbut guessed that it was a screwdriver. McCloskey and Zara­goza devised a test that excluded the misinformation as apossible response alternative, and they found no misinfor­mation effect. Their procedure was analogous to the police­man's asking Mike whether the tool he had seen the robbersteal was "a hammer or a wrench." On the basis of theobservation of reasonably good reporting of the hammer,given this "modified" test, McCloskey and Zaragoza arguedthat it was not necessary to assume any memory impairmentat all-neither impairment oftraces nor impairment ofaccess.

McCloskey and Zaragoza (1985) were right about one thing:When the policeman asked Mike whether he saw a hammeror screwdriver, his answer, "Screwdriver," did not necessarilyimply that he once had a memory for a hammer and thatmemory was now impaired. So the question now is, Doesmisinformation ever produce memory impairment? Belli(1989) says "Maybe," whereas Tversky and Tuchin (1989)say "Yes," on the basis ofnew findings involving "Yes"rNo"retrieval tests, rather than the usual forced-choice recognitiontests. Although the studies are similar in that both involved"Yes"rNo" tests, they differed in terms of the number oftests concerning a critical category of item (e.g., the numberof tests about various tools) that any individual subject re­ceives. More specifically, Belli's procedure is analogous to thepoliceman's asking Mike either "Did you see a hammer-yesor no?'; or "Did you see a wrench-yes or no?" (Le., nosubject received both questions). Tversky and Tuchin's pro­cedure is analogous to the policeman's asking Mike bothquestions and, in addition, a similar question about a screw­driver.

In both studies, the researchers found that misleadingpostevent information reduced the "Yes" responses to thequestion about the original item (e.g., the hammer). In bothstudies, they found that subjects were quite good at rejectingthe novel item (e.g., the wrench). Belli (1989) found thatmisled subjects were better than control subjects at rejectingthe novel item, whereas Tversky and Tuchin (1989) foundthat they were as good as control subjects. Belli never askedpeople what they thought about the misinformation item(e.g., the screwdriver), but Tversky and Tuchin did. Misledsubjects were more likely to incorrectly say that they had seenthe misinformation item than to say that they had seen theone they actually saw. When they adopted the misinformationitem as their own memory, they did so with a high degree ofconfidence, which is not something that one would expectfrom people who are merely guessing.

The collection of experiments seem to be teaching us animportant lesson: When people do not have an original mem­ory, they can and do accept misinformation and adopt it astheir own memory. However, it also appears that misinfor­mation can sometimes impair an otherwise accessible originalmemory. But this conclusion leaves us with many unansweredquestions. How much impairment occurs? What does it mean

to say that memory has been impaired? Is it the memorytraces themselves that are impaired, or is it our ability toreach those memories?

How Much Impairment Occurs?

Belli (1989, Experiment 2) estimated that 32.6% of correctresponses about the event item resulted from an actual mem­ory for the item in the control condition. He estimated that26% ofthe correct responses resulted from an actual memoryfor the item in the misled condition, which was thus a dropof 6.6%. Although the 6.6% difference between control andmisled subjects' performances appeared on the surface to berather small, Belli stressed that the impact of misinformationon actual memory was actually more substantial. When con­sidering only the responses traceable to a true memory for theoriginal item, the misled subjects suffered more than a 20%(6.6% of 32.6%) reduction in accuracy in comparison withcontrol subjects. When looked at in this way, this reductionsuggests that memory impairment can be a significant sourceof erroneous reporting. However, this may not always be thecase. In the experiment in which Belli obtained a 20% im­pairment, the exposure time was 5 s, the interval until mis­information was 5 min, and the final test occurred 10 minafter that. The relative proportion of times that the differentprocess histories cited earlier are responsible for erroneousreporting will depend completely on the conditions of acqui­sition, retention, and retrieval of information. Consider anextreme case: Mike sees a hammer; minutes later Mariamentions a screwdriver, and months pass before the police­man asks Mike, "Did you see a hammer or a screwdriver?"The response of "Screwdriver" would very likely be due topure guessing. On the other hand, if Mike saw the hammer,and months passed before Maria talked to him about ascrewdriver and the policeman questioned him, his responseof "Screwdriver" would very likely be due to what Belli callsmisinformation acceptance.

What Does It Mean to Say ThatMemory Has Been Impaired?

Memory impairment could refer to a weakening ofmemorytraces, or a clouding of memory, or an intrinsic impoverish­ment of memory. It could refer to what an earlier generationof psychologists called "unlearning" (Barnes & Underwood,1959; Melton & Irwin, 1940) or to what a later generationcalled "disintegration" offeatures (Brainerd & Reyna, 1988b).Whatever the mechanism, its fading involves things that wecurrently cannot see or touch but can only infer from behav­ior.

Another potential form of interference has been calledsource misattribution (Lindsay & Johnson, 1987). The ideahere is that there is access to the postevent item but confusionregarding its origin. Although appealing at first glimpse, thenotion of source misattribution has been tossed around; notmuch thought has been given to what it really is in the contextof exposure to misinformation.

Belli (1989) argues that memory impairment and sourcemisattribution hypotheses have one thing in common: the

102 COMMENTS

notion that misleading information interferes with the abilityto re~emberan original item. He combines these hypotheses,refemng to them collectively as misinformation interferencehypotheses, and then distinguishes them from misinformationacceptance hypotheses (in which no interference is assumed).He eventually concludes that some misinformation interfer­ence occurs and then declines to commit to deciding whetherimpairment or source misattribution or both are the respon­sible parties.

This line of argument raises the important question of justwhat source misattribution is. We think that neither Belli(1989) nor Tversky and Tuchin (1989) adequately dealt withthis issue. Source misattribution means, in its most generalsense, confusion over the source of origin of some item. Itmight involve interference with an original item in memory,but on the other hand it might not. Suppose that Mike neversaw the hammer in the first place. He subsequently hearsMaria talk about a screwdriver and decides that he has seenthe screwdriver during the original robbery. Mike has com­mitted a source attribution error, but no interference with anoriginal memory has occurred because there was no originalmemory to begin with. Change the facts slightly and, 10 andbehold, we have on our hands a source attribution error thatis indeed associated with interference with an original item inmemory. If Mike encoded the hammer and subsequentlyencoded "screwdriver" from Maria, he could become con­fused about the source of"screwdriver," could reconstruct hismemory to include a screwdriver, and simultaneously could"impair" his original memory for a hammer. In this case,Mike would have committed a source attribution error thatwas associated with memory impairment. This analysis makesit clear that there is more than one type of source misattri­bution. The type that does not involve memory impairmentis definitely at least partly responsible for the observed results(at least in some subjects); this is in fact what misinformationacceptance is all about.

However, Belli (1989), at least, showed more than this. Heshowed that memory impairment has occurred. It could resultfrom a "clouding" or degrading of memory (picture a Xeroxof a Xerox of a Xerox), or it could be a type of sourcemisattribution that is associated with accessibility of the orig­inal memory, or it could be some ofboth. Whatever he wantsto call it, it appears to involve a type of impairment, and thusBelli's conclusions favoring the memory impairment hypoth­esis could be expressed more strongly than in fact they were.On this fine point, Belli could take a lesson from Tversky andTuchin (1989) on how not to be a shrinking violet. Theyunhesitatingly characterize their data as arguing "against theclaim that nothing happens to the memory for the originalevent as a consequence of misleading information" (p. 89).

In reaching this conclusion, we acknowledge that a differenttype of research is probably necessary before specific claimsabout what "impairment" means can really be addressed.Some biologically oriented psychologists have suggested thatadditional neurobiological facts are required before research­ers can settle the issue of whether impaired performancereflects an actual loss of information from storage and aco~~sponding regression ofsome of the synaptic changes thatongmally represented that stored information (Squire, 1987).

Tversky and Tuchin (1989), in a related vein, asserted thatspecific claims about the retroactive alteration of memorytraces cannot be addressed, given current knowledge and~ools. We bel~eve that the future "knowledge and tools" maymvolve the discovery of new neurobiological facts, but strongadvances in theorizing in this area will occur with new devel­opments in cognitive research.

How Much Guessing Occurs?

Among the reasons why Mike might have reported seeinga screwdriver, one (cited earlier) is that he failed to encodethe hammer and he did not hear Maria mention the screw­driver, but when asked whether it was a hammer or a screw­driver that he saw, he simply guessed that it was a screwdriver.In other words, he made a pure guess, as opposed to a biasedguess. We argued earlier that pure guessing could undercertain conditions of acquisition, retention, and retrieval, beresponsible for a significant proportion of erroneous "screw­driver" reports. However, in the current studies, we believethat pure guessing plays little or no role. Tversky and Tuchin(1989) found that when misled, subjects were certain of theirerrors, which is not the type ofresponse that one would expectfrom people who were merely guessing.

Using the now-familiar burglary sequence (McCloskey &Zaragoza, 1985), Donders, Schooler, and Loftus (1987) alsogathered data that argue against the notion that the "pureguessing" process history makes much of a contribution tothe misinformation effect. In this study, subjects watched theburglary sequence (including, for example, a hammer), thenreceived misinformation (e.g., about a screwdriver) or neutralinformation about four critical items, and then were tested in:h?t Tversky and Tuchin (1989) call the "Loftus test" (e.g.,Old you see a hammer or a screwdriver?"). The innovation

in Donders et at's (1987) research is that speel!ofrespondingwas measured, in addition to confidence. If a high proportionof the misled subjects who selected the misinformation (thescrewdriver) were simply guessing, one would expect theirresponse times to be long and not associated with a highdegree of confidence. The obtained confidence data revealedthat when misled subjects selected the misinformation itemthey did so with a high degree of confidence (just as Tversk;& Tuchin, 1989, found when they used the "Yes"rNo" test).The fastest response times ofall occurred when misled subjectsselected the misinformation item (see Figure 1).

In sum, because subjects embrace the misinformation itemwith a ,high degree ofconfidellce, and they do so very quickly,we believe that pure guessing does not play a significant rolein producing the misinformation effect in studies in whichfairly typical exposure time and retention interval parametersare used. We do not mean to imply that pure guessing neveroccurs, but only that is a rare process history in the misledcondition.

Misinformation Acceptance: A Wcnrthy Phenomenon

Among the many reasons why Mike might have reportedseeing a screwdriver after hearing Maria mention it, one (cited

COMMENTS 103

Figure 1. Reaction times (RT) associated with correct and inco~e~t

responses (Donders, Schooler, & Loftus, 1987). (Msec = mJih­seconds.)

earlier) is that he never saw the hammer in the fIrst place, andhe claimed that he saw the screwdriver because he remem­bered hearing about it. If Mike reported the screwdriver withconviction, we would say, in Belli's (1989) language, that hehad accepted the misinfonnation. In fact, this is one of theclearest cases of misinfonnation acceptance; it cannot bydefInition involve any original memory impainnent becausethere was no original memory to be impaired. Much of thetheoretical discussion about the misinfonnation effect wouldleave us with the impression that this process is uninteresting.It may not, of course, tell us anything about impainnent ofmemories, but it does tell us something about the creation ofnew memories. If a memory for a screwdriver came aboutthrough the process.of suggestion but was subjectively as realand as vivid as a memory that arose from the actual perceptionof a screwdriver, we would fInd this fact important from botha theoretical and an applied perspective. In fact, there isevidence that suggested memories might differ statisticallyfrom genuine memories (Schooler, Gerhard, & Loftus, 1986),but they still have great similarities in common. For onething, suggested memories are expressed with a great deal ofconfIdence, just as are some genuine memories (Donders etaI., 1987; Tversky & Tuchin, 1989). Moreover, suggestedmemories are quickly accessible, just as are some genuinememories (Donders et al., 1987).

Thus researchers in this area should be more interested in"misinfonnation acceptance," especially when it is associatedwith a high degree ofconviction about the new memories. Inexploring the factors that enhance susceptibility to misinfor­mation, we have discoveredl that allowing time to pass afterthe event, so that the original memory can fade, makes aperson particularly vulnerable to suggestion (Loftus, Miller,& Bums, 1978). Some items that could never be modifIedwhen they are fresh in the mind will eventually fade to thepoint that modifIcation is possible. The process could essen­tially involve one of creating a new memory, but it would stillbe an interesting one, worthy of our research attention.

These ideas also bear on Belli's (1989) observation thathaving a better overall memory for original items would makeit easier for misinfonnation interference to be detected (others

uQ)en.sI­a:

3000

2500

2000

1500

1000 -'----'---

Misled ControlCondition

CorrectIncorrect

have similarly argued for the crucial role played by thestrength of the original memory: Brainerd & Reyna, 1988a;Ceci et al., 1987a; Chandler, 1989). Belli did indeed fInd moremisinfonnation interference when there was more memoryto begin with. However, there is a limit. In the extreme case,when memory for an original item is virtually perfect, peopleare unaffected by misinfonnation (e.g., see Loftus, 1979).They readily notice a discrepancy between what they have inmemory and what is being offered to them as misinfonnation(Tousignant, Hall, & Loftus, 1986). In order to affect thosepersons whose memory for some critical item is strong tobegin with, one must wait until the memory fades to a levelbelow which they are not likely to immediately notice dis­crepancies.

Conclusion

We are grateful to McCloskey and Zaragoza (1985) formaking it evident that there are different ways of arriving atthe same memory report. Initially, we thought that memoryimpairment was implicated in the process of distortions inreporting that are due to misleading postevent infonnation.We are gratilled to fInd some support for this notion in thecurrent studies. However, even in the absence of memoryimpainnent, the fInding that people can come to acceptmisinfonnation and adopt it faithfully as their own is animportant phenomenon in its own right. Put another way,regardless of whether there is a buried original memory,waiting to be kissed awake like Sleeping Beauty, researchersstill must take seriously the erroneous memory reports thatare so freely obtained. Researchers have created them inlaboratory environments, which Tversky and Tuchin (1989)claim are "unusual." However, we believe that we have tappeda phenomenon that occurs quite often in real life wheneverpeople who experience the same event talk to one another,overhear each other talk, or gain access to new infonnationfrom the media, interrogators, or other sources. We believethat the misinfonnation effect is sufficiently pervasive andeventually may be so highly controllable that we are temptedto propose a Watsonian future for the misinfonnation effect(see Watson, 1939, p. 104): Give us a dozen healthy memories,well-fonned, and our own specifIed world to handle them in.And we'll guarantee to take anyone at random and train itto become any type of memory that we might select-ham­mer, screwdriver, wrench, stop sign, yield sign, Indian chief­regardless of its origin or the brain that -holds it.

References

A~, R. c., & Shiffrin, R. M. (1968). Human memory: Aproposed system and its control processes. In K. W. Spence &J. T. Spence (Eds.), The psychology of learning and motivation(Vol. 2, pp. 89-195). New York: Academic Press.

Barnes, J. M., & Underwood, B. M. (1959). "Fate" of first-listassociations in transfer theory. Journal ofExperimental Psychology,58.97-105.

Belli, R. F. (1989). Influences of misleading postevent information:Misinformation interference and acceptance. Journal of Experi­mental Psychology: General, 118. 72-85.

104 COMMENTS

Brainerd, C. J., & Reyna, V. F. (I 988a). Memory loci of suggestibilitydevelopment: Comment on Ceci, Ross, and Toglia (1987). JournalofExperimental Psychology: General, 117, 197-200.

Brainerd, C. J., & Reyna, V. F. (l988b). Development offorgettingand reminiscence: A disintegration/redintegration theory. Unpub­lished manuscript, University of Arizona.

Ceci, S. J., Ross, D. F., & Toglia, M. P. (l987a). Age differences insuggestibility: Narrowing the uncertainties. In S. J. Ceci, M. P.Toglia, & D. F. Ross (Eds.), Children's eyewitness testimony (pp.79-91). New York: Springer.

Ceci, S. J., Ross, D. F., & Toglia, M. P. (I 987b). Suggestibility ofchildren's memory: Psycholegal implications. Journal of Experi­mental Psychology: General, 116, 38-49.

Ceci, S. J., Toglia, M. P., & Ross, D. F. (1988). On remembering ...more or less: A trace strength interpretation of developmentaldifferences in suggestibility. Journal of Experimental Psychology:General, 117, 201-203.

Chechile, R. A. (1987). Trace susceptibility theory. Journal ofExper­imental Psychology: General, 116, 203-222.

Chandler, C. C. (1989). Specific retroactive interference in modifiedrecognition tests: Evidence for an unknown cause of interference.Journal ofExperimental Psychology: Learning, Memory, and Cog­nition, 15, 256-265.

Donders, K., Schooler, J. W., & Loftus, E. F. (1987, November).Troubles with memory. Paper presented at the annual meeting ofthe Psychonomic Society, Seattle, Wa.

Geiselman, R. E. (1988). Improving eyewitness memory throughmental reinstatement of context. In G. M. Davies & D. M. Thom­son (Eds.), Memory in context: Context in memory (pp. 245-266).Chichester, England: Wiley.

Gibling, F., & Davies, G. (1988). Reinstatement of context followingexposure to post-event information. British Journal ofPsychology,79, 129-141.

Gudjonsson, G. H. (1986). The relationship between interrogativesuggestibility and acquiescence: Empirical findings and the theo­retical implications. Personality and Individual Differences, 7, 195­199.

Hammersley, R., & Read, J. D. (1986). What is integration? Remem­bering a story and remembering false implications about the story.British Journal ofPsychology, 77, 329-341.

Hammersley, R., & Read, J. D. (in press). What memory changescan account for the misleading question effect? In E. Boyd & L.Radtke (Eds.), Psychological perspectives on eyewitness testimony.New York: Spectrum.

Kohnken, G., & Brockmann, C. (1987). Unspecific postevent infor­mation, attribution of responsibility, and eyewitness performance.Applied Cognitive Psychology, 1, 197-207.

Kroll, N. E. A., & Ogawa, K. H. (1988). Retrieval of the irretrievable:The effect of sequential information on response bias. In M. M.Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects ofmemory: Current research and issues, Vol. 1 (pp. 490-495). Chich­ester, England: Wiley.

Kroll, N. E. A., & Timourian, D. A. (1986). Misleading questionsand the retrieval of the irretrievable. Bulletin of the PsychonomicSociety, 24, 165-168.

Lehnert, W. G. Robertson, S. P., & Black, J. B. (1984). Memoryinteractions during question answering. In J. Mandler, N. L. Stein,& T. Trabasso (Eds.), Learning and comprehension of text. Hills­dale, NJ: Erlbaum.

Lindsay, D. W., & Johnson, M. K. (1987). Reality monitoring andsuggestibility: Children's ability to discriminate among memoriesfrom different sources. In S. J. Ceci, M. P. Toglia, & D. F. Ross

(Eds.), Children's eyewitness memory (chap. 6). New York: Sprin­ger-Verlag.

Loftus, E. F. (1975). Leading questions and the eyewitness report.Cognitive Psychology, 7, 560-572.

Loftus, E. F. (1977). Shifting human color memory. Memory &Cognition, 5, 696-699.

Loftus, E. F. (1979). Eyewitness testimony. Cambridge, MA: HarvardUniversity Press.

Loftus, E. F., Miller, D. G., & Bums, H. J. (1978). Semantic integra­tion of verbal information into a visual memory. Journal ofExper­imental Psychology: Human Learning and Memory, 4, 19-31.

McCloskey, M., & Zaragoza, M. (1985). Misleading postevent infor­mation and memory for events: Arguments and evidence againstmemory impairment hypotheses. Journal of Experimental Psy­chology: General, 114, 1-16.

Melton, A. W., & Irwin, J. M. (1940). The influence of degree ofinterpolated learning on retroactive inhibition and the transfer ofspecific responses. American Journal ofPsychology, 53, 173-203.

Morton, J., Hammersley, R. H., & Bekerian, D. A. (1985). Headedrecords: A model for memory and its failures. Cognition, 20, 1-23.

Pirolli, P. L., & Mitterer, J. O. (1984). The effect ofleading questionson prior memory: Evidence for the coexistence of inconsistentmemory traces. Canadian Journal ofPsychology, 38, 135-14 I.

Register, P. A., & Kihlstrom, J. F. (1988). Hypnosis and interrogativesuggestibility. Personality and Individual Differences, 9, 549-558.

Schooler, J. W., Gerhard, D., & Loftus, E. F. (1986). Qualities of theunreal. Journal of Experimental Psychology: Learning, Memory,and Cognition, 12, 171-18 I.

Sheehan, P. W. (1988). Confidence, memory and hypnosis. In H.Pettinati (Ed.), Hypnosis and memory (pp. 96-154). New York:Guilford.

Sheehan, P. W., & Tilden, J. (1986). The consistency of occurrencesof memory distortion following hypnotic induction. InternationalJournal ofClinical and Experimental Hypnosis, 34, 122-137.

Smith, V. L., & Ellsworth, P. C. (1987). The social psychology ofeyewitness accuracy: Misleading questions and communicator ex­pertise. Journal ofApplied Psychology, 72, 294-300.

Squire, L. R. (1987). Memory and brain. Oxford, England: OxfordUniversity Press.

Tousignant, J. P., Hall, D., & Loftus, E. F. (1986). Discrepancydetection and vulnerability to misleading post-event information.Memory & Cognition, 14, 329-338.

Tversky, B., & Tuchin, M. (1989). A reconciliation of the evidenceon eyewitness testimony: Comments on McCloskey and Zaragoza(1985). Journal ofExperimental Psychology: General, 118, 86-91.

Wagenaar, W. A., & Boer, H. P. A. (1987). Misleading posteventinformation: Testing parameterized models of integration in mem­ory. Acta Psychologica, 66,291-306.

Watson, J. B. (1939). Behaviorism (2nd ed). Chicago: University ofChicago Press.

Zaragoza, M. S., & Koshmider, J. W. (1989). Misled subjects mayknow more than their performance implies. Journal ofExperimen­tal Psychology: Learning, Memory, and Cognition, 15, 246-255.

Zaragoza, M. S., McCloskey, M., & Jamis, M. (1987). Misleadingpostevent information and recall of the original event: Furtherevidence against the memory impairment hypothesis. Journal ofExperimental Psychology: Learning, Memory, and Cognition, 13,36-44.

Received September 13, 1988Accepted September 19, 1988 •